Simplified quantification of dopamine transporters in humans using [99mTc]TRODAT-1 and single-photon emission tomography

Abstract. Quantification of dopamine transporters (DAT) using [99mTc]TRODAT-1 and single-photon emission tomography (SPET) requires full kinetic modeling of the data, using complex and invasive arterial blood sampling to provide an input function to the model. We have shown previously that a simpler reference tissue model provides accurate quantitative results, using a reference region devoid of DAT as the input to the model and thereby obviating the need for blood sampling. We now extend this work into humans, and develop further simplifications to make the imaging protocol much more practical as a routine procedure. Fourteen healthy subjects (age 29.8±8.4 years, range 18.7–45.5 years) underwent dynamic SPET for 6 h following injection of 752±28 MBq [99mTc]TRODAT-1. The kinetic data were analyzed using nonlinear regression analysis (NLRA) and Logan-Patlak graphical analysis. In addition, simple average ratios of striatal-to-background counts were obtained for three 1-h periods (3–4 h, 4–5 h, 5–6 h), and compared against the kinetic models. All methods gave an index of specific binding, proportional to the binding potential, known as the distribution volume ratio (DVR). The reference tissue NLRA gave mean values of k3=0.013±0.003 min–1, k4=0.011±0.002 min–1, and DVR=2.29±0.17. Graphical analysis gave a value of DVR=2.28±0.16, and the three ratio values of DVR were: 3–4 h, 2.18±0.15; 4–5 h, 2.34±0.13; and 5–6 h, 2.46±0.19. Graphical analysis was highly correlated with NLRA (R2=0.91, slope=0.90±0.08). The ratio methods correlated well with NLRA (3–4 h, R2=0.71, slope=0.73±0.13; 4–5 h, R2=0.86, slope=0.73±0.09; 5–6 h, R2=0.80, slope=1.00±0.15), and also with graphical analysis (3–4 h, R2=0.65, slope=0.74±0.16; 4–5 h, R2=0.85, slope=0.78±0.09; 5–6 h, R2=0.88, slope=1.11±0.12). The optimum equilibrium time point for obtaining a simple ratio was approximately 4.5–5.5 h. In conclusion, the simple ratio techniques for obtaining a quantitative measure of specific binding correlated well with the reference tissue kinetic models, using both NLRA and graphical analysis. The optimum time for obtaining a ratio appeared to be in the range 4.5–5.5 h. Earlier time points, while still relatively accurate, had a lower sensitivity and may not be optimized for measuring small changes in DAT concentrations.